linux-next.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Coverity: wilc_parse_join_bss_param(): Memory - illegal accesses
@ 2019-11-04 17:38 coverity-bot
  2019-11-05  6:11 ` Ajay.Kathat
  0 siblings, 1 reply; 4+ messages in thread
From: coverity-bot @ 2019-11-04 17:38 UTC (permalink / raw)
  To: Ajay Singh; +Cc: Greg Kroah-Hartman, Gustavo A. R. Silva, linux-next

Hello!

This is an experimental automated report about issues detected by Coverity
from a scan of next-20191031 as part of the linux-next weekly scan project:
https://scan.coverity.com/projects/linux-next-weekly-scan

You're getting this email because you were associated with the identified
lines of code (noted below) that were touched by recent commits:

4e0b0f42c9c7 ("staging: wilc1000: use struct to pack join parameters for FW")

Coverity reported the following:

*** CID 1487400:  Memory - illegal accesses  (OVERRUN)
/drivers/staging/wilc1000/wilc_hif.c: 496 in wilc_parse_join_bss_param()
490     	if (supp_rates_ie) {
491     		if (supp_rates_ie[1] > (WILC_MAX_RATES_SUPPORTED - rates_len))
492     			param->supp_rates[0] = WILC_MAX_RATES_SUPPORTED;
493     		else
494     			param->supp_rates[0] += supp_rates_ie[1];
495
vvv     CID 1487400:  Memory - illegal accesses  (OVERRUN)
vvv     Overrunning array of 13 bytes at byte offset 13 by dereferencing pointer "&param->supp_rates[rates_len + 1]". [Note: The source code implementation of the function has been overridden by a builtin model.]
496     		memcpy(&param->supp_rates[rates_len + 1], supp_rates_ie + 2,
497     		       (param->supp_rates[0] - rates_len));
498     	}
499
500     	ht_ie = cfg80211_find_ie(WLAN_EID_HT_CAPABILITY, ies->data, ies->len);
501     	if (ht_ie)

If this is a false positive, please let us know so we can mark it as
such, or teach the Coverity rules to be smarter. If not, please make
sure fixes get into linux-next. :) For patches fixing this, please
include these lines (but double-check the "Fixes" first):

Reported-by: coverity-bot <keescook+coverity-bot@chromium.org>
Addresses-Coverity-ID: 1487400 ("Memory - illegal accesses")
Fixes: 4e0b0f42c9c7 ("staging: wilc1000: use struct to pack join parameters for FW")


Thanks for your attention!

-- 
Coverity-bot

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Coverity: wilc_parse_join_bss_param(): Memory - illegal accesses
  2019-11-04 17:38 Coverity: wilc_parse_join_bss_param(): Memory - illegal accesses coverity-bot
@ 2019-11-05  6:11 ` Ajay.Kathat
  2019-11-05 17:12   ` Kees Cook
  0 siblings, 1 reply; 4+ messages in thread
From: Ajay.Kathat @ 2019-11-05  6:11 UTC (permalink / raw)
  To: keescook; +Cc: gregkh, gustavo, linux-next, Adham.Abozaeid



On 04-Nov-19 11:08 PM, coverity-bot wrote:
> External E-Mail
> 
> 
> Hello!
> 
> This is an experimental automated report about issues detected by Coverity
> from a scan of next-20191031 as part of the linux-next weekly scan project:
> https://scan.coverity.com/projects/linux-next-weekly-scan
> 
> You're getting this email because you were associated with the identified
> lines of code (noted below) that were touched by recent commits:
> 
> 4e0b0f42c9c7 ("staging: wilc1000: use struct to pack join parameters for FW")
> 
> Coverity reported the following:
> 
> *** CID 1487400:  Memory - illegal accesses  (OVERRUN)
> /drivers/staging/wilc1000/wilc_hif.c: 496 in wilc_parse_join_bss_param()
> 490     	if (supp_rates_ie) {
> 491     		if (supp_rates_ie[1] > (WILC_MAX_RATES_SUPPORTED - rates_len))
> 492     			param->supp_rates[0] = WILC_MAX_RATES_SUPPORTED;
> 493     		else
> 494     			param->supp_rates[0] += supp_rates_ie[1];
> 495
> vvv     CID 1487400:  Memory - illegal accesses  (OVERRUN)
> vvv     Overrunning array of 13 bytes at byte offset 13 by dereferencing pointer "&param->supp_rates[rates_len + 1]". [Note: The source code implementation of the function has been overridden by a builtin model.]
> 496     		memcpy(&param->supp_rates[rates_len + 1], supp_rates_ie + 2,
> 497     		       (param->supp_rates[0] - rates_len));

As I understand, Ideally this condition should never arise because the
maximum number of supported *basic rates* is up to 8 so the value of
‘rate_len’ will always be less then WILC_MAX_RATES_SUPPPRTED (i.e 12).
Therefore '&param->supp_rates[rates_len+ 1]' will never try to access
the 13 bytes in the array.
But for the safer side, if need I can create a patch to block the
addition of extended supported rates in ‘param->supp_rates’ array if
‘rates_len’ is 12 (i.e 'param->supp_rates' array is full after filing
the basic supported rates).

> 498     	}
> 499
> 500     	ht_ie = cfg80211_find_ie(WLAN_EID_HT_CAPABILITY, ies->data, ies->len);
> 501     	if (ht_ie)
> 
> If this is a false positive, please let us know so we can mark it as
> such, or teach the Coverity rules to be smarter. If not, please make
> sure fixes get into linux-next. :) For patches fixing this, please
> include these lines (but double-check the "Fixes" first):
> 
> Reported-by: coverity-bot <keescook+coverity-bot@chromium.org>
> Addresses-Coverity-ID: 1487400 ("Memory - illegal accesses")
> Fixes: 4e0b0f42c9c7 ("staging: wilc1000: use struct to pack join parameters for FW")
> 
> 
> Thanks for your attention!
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Coverity: wilc_parse_join_bss_param(): Memory - illegal accesses
  2019-11-05  6:11 ` Ajay.Kathat
@ 2019-11-05 17:12   ` Kees Cook
  2019-11-06  5:13     ` Ajay.Kathat
  0 siblings, 1 reply; 4+ messages in thread
From: Kees Cook @ 2019-11-05 17:12 UTC (permalink / raw)
  To: Ajay.Kathat; +Cc: gregkh, gustavo, linux-next, Adham.Abozaeid

On Tue, Nov 05, 2019 at 06:11:03AM +0000, Ajay.Kathat@microchip.com wrote:
> 
> 
> On 04-Nov-19 11:08 PM, coverity-bot wrote:
> > External E-Mail
> > 
> > 
> > Hello!
> > 
> > This is an experimental automated report about issues detected by Coverity
> > from a scan of next-20191031 as part of the linux-next weekly scan project:
> > https://scan.coverity.com/projects/linux-next-weekly-scan
> > 
> > You're getting this email because you were associated with the identified
> > lines of code (noted below) that were touched by recent commits:
> > 
> > 4e0b0f42c9c7 ("staging: wilc1000: use struct to pack join parameters for FW")
> > 
> > Coverity reported the following:
> > 
> > *** CID 1487400:  Memory - illegal accesses  (OVERRUN)
> > /drivers/staging/wilc1000/wilc_hif.c: 496 in wilc_parse_join_bss_param()
> > 490     	if (supp_rates_ie) {
> > 491     		if (supp_rates_ie[1] > (WILC_MAX_RATES_SUPPORTED - rates_len))
> > 492     			param->supp_rates[0] = WILC_MAX_RATES_SUPPORTED;
> > 493     		else
> > 494     			param->supp_rates[0] += supp_rates_ie[1];
> > 495
> > vvv     CID 1487400:  Memory - illegal accesses  (OVERRUN)
> > vvv     Overrunning array of 13 bytes at byte offset 13 by dereferencing pointer "&param->supp_rates[rates_len + 1]". [Note: The source code implementation of the function has been overridden by a builtin model.]
> > 496     		memcpy(&param->supp_rates[rates_len + 1], supp_rates_ie + 2,
> > 497     		       (param->supp_rates[0] - rates_len));
> 
> As I understand, Ideally this condition should never arise because the
> maximum number of supported *basic rates* is up to 8 so the value of
> ‘rate_len’ will always be less then WILC_MAX_RATES_SUPPPRTED (i.e 12).
> Therefore '&param->supp_rates[rates_len+ 1]' will never try to access
> the 13 bytes in the array.
> But for the safer side, if need I can create a patch to block the
> addition of extended supported rates in ‘param->supp_rates’ array if
> ‘rates_len’ is 12 (i.e 'param->supp_rates' array is full after filing
> the basic supported rates).

I don't know the code myself, but generally speaking, it's best to
validate any assumptions like this. I'd say add a patch for it, since it
sounds pretty straight-forward to test.

-Kees

> 
> > 498     	}
> > 499
> > 500     	ht_ie = cfg80211_find_ie(WLAN_EID_HT_CAPABILITY, ies->data, ies->len);
> > 501     	if (ht_ie)
> > 
> > If this is a false positive, please let us know so we can mark it as
> > such, or teach the Coverity rules to be smarter. If not, please make
> > sure fixes get into linux-next. :) For patches fixing this, please
> > include these lines (but double-check the "Fixes" first):
> > 
> > Reported-by: coverity-bot <keescook+coverity-bot@chromium.org>
> > Addresses-Coverity-ID: 1487400 ("Memory - illegal accesses")
> > Fixes: 4e0b0f42c9c7 ("staging: wilc1000: use struct to pack join parameters for FW")
> > 
> > 
> > Thanks for your attention!
> > 

-- 
Kees Cook

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Coverity: wilc_parse_join_bss_param(): Memory - illegal accesses
  2019-11-05 17:12   ` Kees Cook
@ 2019-11-06  5:13     ` Ajay.Kathat
  0 siblings, 0 replies; 4+ messages in thread
From: Ajay.Kathat @ 2019-11-06  5:13 UTC (permalink / raw)
  To: keescook; +Cc: gregkh, gustavo, linux-next, Adham.Abozaeid



On 05-Nov-19 10:42 PM, Kees Cook wrote:
> External E-Mail
> 
> 
> On Tue, Nov 05, 2019 at 06:11:03AM +0000, Ajay.Kathat@microchip.com wrote:
>>
>>
>> On 04-Nov-19 11:08 PM, coverity-bot wrote:
>>> External E-Mail
>>>
>>>
>>> Hello!
>>>
>>> This is an experimental automated report about issues detected by Coverity
>>> from a scan of next-20191031 as part of the linux-next weekly scan project:
>>> https://scan.coverity.com/projects/linux-next-weekly-scan
>>>
>>> You're getting this email because you were associated with the identified
>>> lines of code (noted below) that were touched by recent commits:
>>>
>>> 4e0b0f42c9c7 ("staging: wilc1000: use struct to pack join parameters for FW")
>>>
>>> Coverity reported the following:
>>>
>>> *** CID 1487400:  Memory - illegal accesses  (OVERRUN)
>>> /drivers/staging/wilc1000/wilc_hif.c: 496 in wilc_parse_join_bss_param()
>>> 490     	if (supp_rates_ie) {
>>> 491     		if (supp_rates_ie[1] > (WILC_MAX_RATES_SUPPORTED - rates_len))
>>> 492     			param->supp_rates[0] = WILC_MAX_RATES_SUPPORTED;
>>> 493     		else
>>> 494     			param->supp_rates[0] += supp_rates_ie[1];
>>> 495
>>> vvv     CID 1487400:  Memory - illegal accesses  (OVERRUN)
>>> vvv     Overrunning array of 13 bytes at byte offset 13 by dereferencing pointer "&param->supp_rates[rates_len + 1]". [Note: The source code implementation of the function has been overridden by a builtin model.]
>>> 496     		memcpy(&param->supp_rates[rates_len + 1], supp_rates_ie + 2,
>>> 497     		       (param->supp_rates[0] - rates_len));
>>
>> As I understand, Ideally this condition should never arise because the
>> maximum number of supported *basic rates* is up to 8 so the value of
>> ‘rate_len’ will always be less then WILC_MAX_RATES_SUPPPRTED (i.e 12).
>> Therefore '&param->supp_rates[rates_len+ 1]' will never try to access
>> the 13 bytes in the array.
>> But for the safer side, if need I can create a patch to block the
>> addition of extended supported rates in ‘param->supp_rates’ array if
>> ‘rates_len’ is 12 (i.e 'param->supp_rates' array is full after filing
>> the basic supported rates).
> 
> I don't know the code myself, but generally speaking, it's best to
> validate any assumptions like this. I'd say add a patch for it, since it
> sounds pretty straight-forward to test.

Sure, I will add a patch for this.

Regards,
Ajay

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2019-11-06  5:13 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-11-04 17:38 Coverity: wilc_parse_join_bss_param(): Memory - illegal accesses coverity-bot
2019-11-05  6:11 ` Ajay.Kathat
2019-11-05 17:12   ` Kees Cook
2019-11-06  5:13     ` Ajay.Kathat

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).